{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:5VJO4WULHKPUAIP7IS6BAPLFKW","short_pith_number":"pith:5VJO4WUL","canonical_record":{"source":{"id":"2502.13928","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-02-19T18:05:42Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"7894caf675b69efc089c74f1b077020abdaabfb1b35fae7b9f0a6b54c4822e01","abstract_canon_sha256":"beb2445f5a685c129d39f109985238de193e592a05918812af8c6cc60e3fe3ae"},"schema_version":"1.0"},"canonical_sha256":"ed52ee5a8b3a9f4021ff44bc103d6555a805aa00925e3580151000f260176533","source":{"kind":"arxiv","id":"2502.13928","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.13928","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"arxiv_version","alias_value":"2502.13928v2","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.13928","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"pith_short_12","alias_value":"5VJO4WULHKPU","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"pith_short_16","alias_value":"5VJO4WULHKPUAIP7","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"pith_short_8","alias_value":"5VJO4WUL","created_at":"2026-07-05T11:14:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:5VJO4WULHKPUAIP7IS6BAPLFKW","target":"record","payload":{"canonical_record":{"source":{"id":"2502.13928","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-02-19T18:05:42Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"7894caf675b69efc089c74f1b077020abdaabfb1b35fae7b9f0a6b54c4822e01","abstract_canon_sha256":"beb2445f5a685c129d39f109985238de193e592a05918812af8c6cc60e3fe3ae"},"schema_version":"1.0"},"canonical_sha256":"ed52ee5a8b3a9f4021ff44bc103d6555a805aa00925e3580151000f260176533","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:27.538186Z","signature_b64":"Qkx6Jq7SCxx51opsOhUVIq3Eu1hvqp/x6IAsz4AAFSIcGHrF8eelXAFcuCJwueerKtdqIPZ4hV1bEGAfX1uWCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ed52ee5a8b3a9f4021ff44bc103d6555a805aa00925e3580151000f260176533","last_reissued_at":"2026-07-05T11:14:27.537664Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:27.537664Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.13928","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:14:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UsPVx37BeJeQIsjlhzhcykMlN79tJiojVJl9NP9yU3OMvUIzroCrkmtrYGJjkAqSfA5H/LLpvH0WFBierCLvAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T06:24:27.301232Z"},"content_sha256":"80e0327228e8e6d5f4866fb6ee886cca62f804f70949d1c0d9f206bcd31133fd","schema_version":"1.0","event_id":"sha256:80e0327228e8e6d5f4866fb6ee886cca62f804f70949d1c0d9f206bcd31133fd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:5VJO4WULHKPUAIP7IS6BAPLFKW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Symmetrical Visual Contrastive Optimization: Aligning Vision-Language Models with Minimal Contrastive Images","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Fan-Yun Sun, Kaiyue Wen, Nick Haber, Shengguang Wu","submitted_at":"2025-02-19T18:05:42Z","abstract_excerpt":"Recent studies have shown that Large Vision-Language Models (VLMs) tend to neglect image content and over-rely on language-model priors, resulting in errors in visually grounded tasks and hallucinations. We hypothesize that this issue arises because existing VLMs are not explicitly trained to generate texts that are accurately grounded in fine-grained image details. To enhance visual feedback during VLM training, we propose S-VCO (Symmetrical Visual Contrastive Optimization), a novel finetuning objective that steers the model toward capturing important visual details and aligning them with cor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.13928","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2502.13928/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:14:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aPYRDa/78fhZs3KbuvZMv6sOUAHJkxW7AHiHBaPMW98DbdfjBxwNkBjIDVihDYKwJaI2yFDLLmVq9zqczZu6Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T06:24:27.301809Z"},"content_sha256":"19d947c78cdb1e284386fe6f02c66809564d1c6f40a292f5ce7f85df94169d0a","schema_version":"1.0","event_id":"sha256:19d947c78cdb1e284386fe6f02c66809564d1c6f40a292f5ce7f85df94169d0a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5VJO4WULHKPUAIP7IS6BAPLFKW/bundle.json","state_url":"https://pith.science/pith/5VJO4WULHKPUAIP7IS6BAPLFKW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5VJO4WULHKPUAIP7IS6BAPLFKW/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-08T06:24:27Z","links":{"resolver":"https://pith.science/pith/5VJO4WULHKPUAIP7IS6BAPLFKW","bundle":"https://pith.science/pith/5VJO4WULHKPUAIP7IS6BAPLFKW/bundle.json","state":"https://pith.science/pith/5VJO4WULHKPUAIP7IS6BAPLFKW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5VJO4WULHKPUAIP7IS6BAPLFKW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:5VJO4WULHKPUAIP7IS6BAPLFKW","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"beb2445f5a685c129d39f109985238de193e592a05918812af8c6cc60e3fe3ae","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-02-19T18:05:42Z","title_canon_sha256":"7894caf675b69efc089c74f1b077020abdaabfb1b35fae7b9f0a6b54c4822e01"},"schema_version":"1.0","source":{"id":"2502.13928","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.13928","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"arxiv_version","alias_value":"2502.13928v2","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.13928","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"pith_short_12","alias_value":"5VJO4WULHKPU","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"pith_short_16","alias_value":"5VJO4WULHKPUAIP7","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"pith_short_8","alias_value":"5VJO4WUL","created_at":"2026-07-05T11:14:27Z"}],"graph_snapshots":[{"event_id":"sha256:19d947c78cdb1e284386fe6f02c66809564d1c6f40a292f5ce7f85df94169d0a","target":"graph","created_at":"2026-07-05T11:14:27Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2502.13928/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent studies have shown that Large Vision-Language Models (VLMs) tend to neglect image content and over-rely on language-model priors, resulting in errors in visually grounded tasks and hallucinations. We hypothesize that this issue arises because existing VLMs are not explicitly trained to generate texts that are accurately grounded in fine-grained image details. To enhance visual feedback during VLM training, we propose S-VCO (Symmetrical Visual Contrastive Optimization), a novel finetuning objective that steers the model toward capturing important visual details and aligning them with cor","authors_text":"Fan-Yun Sun, Kaiyue Wen, Nick Haber, Shengguang Wu","cross_cats":["cs.AI","cs.CL","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-02-19T18:05:42Z","title":"Symmetrical Visual Contrastive Optimization: Aligning Vision-Language Models with Minimal Contrastive Images"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.13928","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:80e0327228e8e6d5f4866fb6ee886cca62f804f70949d1c0d9f206bcd31133fd","target":"record","created_at":"2026-07-05T11:14:27Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"beb2445f5a685c129d39f109985238de193e592a05918812af8c6cc60e3fe3ae","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-02-19T18:05:42Z","title_canon_sha256":"7894caf675b69efc089c74f1b077020abdaabfb1b35fae7b9f0a6b54c4822e01"},"schema_version":"1.0","source":{"id":"2502.13928","kind":"arxiv","version":2}},"canonical_sha256":"ed52ee5a8b3a9f4021ff44bc103d6555a805aa00925e3580151000f260176533","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ed52ee5a8b3a9f4021ff44bc103d6555a805aa00925e3580151000f260176533","first_computed_at":"2026-07-05T11:14:27.537664Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:14:27.537664Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Qkx6Jq7SCxx51opsOhUVIq3Eu1hvqp/x6IAsz4AAFSIcGHrF8eelXAFcuCJwueerKtdqIPZ4hV1bEGAfX1uWCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:14:27.538186Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.13928","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:80e0327228e8e6d5f4866fb6ee886cca62f804f70949d1c0d9f206bcd31133fd","sha256:19d947c78cdb1e284386fe6f02c66809564d1c6f40a292f5ce7f85df94169d0a"],"state_sha256":"f19e4dd4d71c4c746af0987ea97cdf52b335f77e2d270b25c85efb290da4b891"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SYBeUtk5xZP3SjHZOXuyM72O1C/ZVJNGE9JC7n6H/T6D9xA1GLypHXB9moa9iHMTmeEnuBsQntBQpFUPrBuTBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T06:24:27.307466Z","bundle_sha256":"4d38a584049c3eb7f772b53b32f8d7142469cc724757135ea16be4e1b55f598c"}}